模糊运动图像的异常动作轨迹矢量搜索方法

Vector searching method for abnormal motion trajectory of blurred moving images

为了解决当前模糊图像异常动作轨迹的搜索方法忽略模糊运动图像成因,造成的图像复原不完整、效果差及搜索效率低的问题,提出模糊运动图像的异常动作轨迹矢量搜索方法.采用最小二乘法对模糊运动图像进行复原并提取复原图像中存在的关键像素点构建运动图像的样本库,再运用BDT算法结合图像样本库搜索异常动作轨迹矢量,完成模糊运动图像中异常动作轨迹矢量的搜索.实验结果表明,所提方法的图像复原效果好、搜索效率高.为相关的模糊图像异常动作轨迹的搜索方法提供参考.

In order to solve the problems of incomplete image restoration,poor effect and low search efficiency caused by the ignorance of fuzzy motion image origination in current searching methods for abnormal motion trajectory of fuzzy image,a vector searching method for abnormal motion trajectory of fuzzy motion image was proposed.The least square method was used to restore the blurred motion image,and the key pixels in the restored image are extracted to construct the sample library of moving image.Thereafter,the BDT algorithm combined with the image sample library was used to search the abnormal motion trajectory vector of the blurred moving image.Experimental results show that the as-proposed method has better image restoration effect and higher searching efficiency.It provides a reference for the searching for abnormal motion trajectory of fuzzy image.